An Improved Particle Swarm Optimization for the Constrained Portfolio Selection Problem

We focus on a constrained portfolio selection model with transaction costs and quantity limit. Due to these complex constraints, the process becomes a high-dimensional constrained optimization problem. Traditional optimization algorithms fail to work efficiently and heuristic algorithms with effective searching ability can be the best choose for the problem, and then we design an improved particle swarm (ISPO) optimization to solve this question. In order to prevent premature convergence to local minima, we design a new definition for global point. Finally, a numerical example of a portfolio selection problem is given to illustrate our proposed method; the simulation results demonstrate good performance of the IPSO in solving the complex constrained portfolio selection problem.